NATIOMEM Reduced preferential sputtering of TiO 2 (and Ta 2 O 5 ) thin films through argon cluster ion bombardment. R. Grilli *, P. Mack, M.A. Baker * * University of Surrey, UK ThermoFisher Scientific Ltd, UK 3 rd Vacuum Symposium UK 18 October 2012 Coventry, UK
Contents: - NATIOMEM project - Introduction: preferential sputtering - Experimental details - XPS surface characterization - Etching results - Interpretation of oxygen peaks - Ti 2p quantification - Conclusions
NATIOMEM EU project NAno-structured TION Photo-Catalytic MEMbranes for Water Treatment Overlall objective: develop a nanostructured catalytic membrane that will be a cost effective component in solar radiation based treatment systems for converting surface or waste water into drinking water using membrane substrates coated with nanostructured photocatalytic N-doped TiO 2 thin films. Photocatalysis Microbial contaminants: Protozoa, bacteria, virus. Other contaminants: TOC, colour, taste, odor substances, pollutants (agro and industrial chemicals, pharmaceuticals, heavy metals) UV/vis OH virus r Porous membrane N-doped TiO 2 coating
XPS analysis employed to accurately determine N-doped TiO 2 thin film composition (sol-gel, FVAD, sputtering) Contaminant species at surface interfere with the analysis and are removed through Ar + ion sputtering prior to XPS spectral acquisition Hydrocarbons, hydroxides, adsorbed H 2 O N-doped TiO 2 film deposited on alumina membrane by sol-gel N-TiO 2
Preferential sputtering Preferential sputtering One component sputtered in preference to the other(s) - Surface binding effects - Mass difference effects Affected oxides: Preferential removal of O atoms Ti, Ta, Co, Ni, Cu, Nb, Ru, Ag, W, Pb, Bi, Hf Ion source Alternatives Ar + Ar n+, C 60+, SF 5+, C 24 H 12 +
Monomer vs Cluster damage Ga at 15 kev C 60 at 15 kev Ga + + Tt=29 ps Microscopic insights into the sputtering of Ag{111} induced by C 60 and Ga Bombardment, J. Phys. Chem. B, 108, 7831-7838 (2004).
Effects of sputtering TiO 2 TiO 2 TiO x x < 2 TiO 2 reduces to Ti 2 O 3 and TiO Before etching Ti 2p After etching TiO 2 Ti 4+ Ti 2 O 3 Ti 3+ TiO Ti 2+ BE = 458.8 ev BE = 457.8 ev BE = 455.6 ev 468 466 464 462 460 458 456 468 466 464 462 460 458 456 454 452 Bonding and quantitative information on the original film is lost
Experimental details Theta Probe, by Thermo Scientific 1. Surface characterisation: Source: Monocromated Al Kα, (hν=1486.6 ev) 140W, 400 μm spot Pass energy: 300 ev for survey 50 ev for high resolution spectra K-Alpha, by Thermo Scientific 2. Etching at 3 kv and 3. 200 V: Source: Al Kα, (hν=1486.6 ev) Ion gun: Ar + monomer Etching 1 Voltage: 3 kv, Beam current: 4 μa, Raster size: 2 mm 4 mm Etching 2 Voltage: 200 V, Beam current: 1 μa, Raster size: 2 mm 4 mm ESCALAB Mk II, by Thermo Scientific 4. Etching with 4 kv Ar n + cluster source: Source: Al Kα, (hν=1486.6 ev) Ion gun: Ar n + clusters (prototype) Cluster size: ~ 1000 atoms Etching 3 Voltage: 4 kv, Beam current: 10 na, Raster size: 2 mm 2 mm Adventitious C-H (C1s = 285.0 ev) was used as reference for charge shift correction Avantage software (Thermo Scientific) used for data processing
XPS surface characterization Specimen: Undoped nominal TiO 2 thin film deposited by reactive FVAD O 1s Ti 2p C KLL Ti LMM O KLL Ti 2s C 1s N 1s Ti 3s Ti 3p
Composition of the native TiO 2 film surface O 1s O oxide Ti 2p Ti 4+ (TiO 2 ) TiO 2 42.5% 530.4 ev Ti 4+ 18.4% 458.9 ev -OH 3.2% 531.5 ev Ti 3+ 0.6% 457.4 ev -OH H 2 O 2.9% 532.4 ev H 2 O Ti 3+ (defect) 537 535 533 531 529 527 525 468 466 464 462 460 458 456 454 452 450 Composition (at%) C1s Ti2p N1s O1s tot O1s ox O tot /Ti O ox /Ti 31.5 19.0 0.9 48.6 42.5 2.56 2.24
Composition (at%) Etching results: Ar + ion beam, 3 kv 3 kv 70 60 O1s 50 40 30 Ti O C Oex 20 10 536 534 532 530 528 526 0 0 5 10 15 20 25 30 35 40 45 Time (s) Ti2p TiO 2 Ti 2 O 3 C1s Ti2p O1s tot O1s ox O tot /Ti O ox /Ti Comp.(at%) - 30.5 69.5 54.4 2.27 1.78 TiO O oxide 52.5% 530.4 ev Ti 4+ 9.7% 458.8 ev -OH 10.2% 531.9 ev H 2 O 6.4% 533.3 ev Ti 3+ 13.2% 457.3 ev Ti 2+ 7.8% 455.7 ev 468 464 460 456 452 Charge shift correction: no
Composition (at%) Etching results: Ar + ion beam, 200 V 200 ev etch 70 60 O1s 50 40 30 Titot Oox C Odef 20 10 0 0 10 20 30 40 50 60 70 80 90 Time (s) 538 536 Ti2p 534 532 530 528 TiO 2 Ti 2 O 3 526 C1s Ti2p O1s tot O1s ox O tot /Ti O ox /Ti Comp.(at%) 0.5 30.8 68.7 53.6 2.23 1.74 TiO O oxide 53.5% 530.4 ev Ti 4+ 14.6% 458.7 ev -OH 11.6% 531.9 ev Ti 3+ 10.7% 457.0 ev H 2 O 3.5% 533.3 ev Ti 2+ 4.9% 455.4 ev Charge shift correction: no 468 466 464 462 460 458 456 454 452
0 60 120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 Composition (at%) Etching results: Ar + cluster source 4 kv Cluster Source 80 70 60 50 40 30 20 10 Ti O C Oex O1s N 535 534 533 532 531 530 529 528 0 Time (s) Ti2p TiO 2 C1s Ti2p O1s tot O1s ox O tot /Ti O ox /Ti Comp.(at%) 3.4 27.9 68.7 64.8 2.46 2.32 Ti 2 O 3 O oxide 64.5% 531.0 ev -OH 3.9% 532.4 ev Ti 4+ 24.4% 459.5 ev Ti 3+ 3.14% 457.8 ev 475 470 465 460 455 450 Charge shift correction: yes
Oxygen peak interpretation Oxides (TiO 2, Ti 2 O 3, TiO) H 2 O? -OH? Surface hydroxyl groups and adsorbed water should be only on the surface, but the peaks at higher BE increase with etch time. Possible interpretations from literature: 1. Back-deposition of oxygen on the surface 2. After reduction of Ti 4+ to Ti 3+ and Ti 2+ the excess O diffuses from the bulk towards the surface and forms OH groups 3. Free electrons in the film may become localized at Ti sites, forming Ti 3+ and increasing the Ti-O bond length, because of partial filling of anti-bonding orbitals. This may increase the BE of O1s electrons. 4. Formation of defective oxygen sites: highly polarized oxygen atoms near to oxygen vacancies (z = +2) or tri/tetravalent Ti interstitials (z = +3, +4) 5. Oxygen atoms with a lower electron density, described as O -, this happens when the co-ordination number is less than 3 (less than 3 Ti atoms bonded to one O), giving M-O bonds with a higher covalent nature. The atoms may re-arrange to this configuration after formation of a Ti vacancy.
3 kv etch 2p1/2 loss 2p3/2 2p3/2 Ti2p quantification issues Quantification of TiO 2 always shows an excess of oxygen 200 V etch 2p1/2 loss 13 ev Ti 2p wider BE acquisition range 3 ev 480 475 470 465 460 455 M.Oku et al. Journal of Electron Spectroscopy and Related Phenomena, 105 (1999) 211-218
O/Ti O/Ti ratio after sputtering for the different ion beam conditions O/Ti ratios 2.6 2.4 2.2 2 1.8 3 kv 200 V Cluster 1.6 1.4 0 10 20 30 40 50 60 70 80 90 Time (s) 150 210 270 330 390 450
Tantalum oxide film (Ta 2 O 5 ) Even low energy monomer cleaning causes a significant amount of reduction Argon cluster cleaning gives no visible sign of oxide reduction Quantification after argon cluster cleaning confirms Ta 2 O 5 stoichiometry (± 1 at%) Clean method Ta4f oxide Ta4f reduced None 100 - Cluster 100-200eV monomer 70.4 29.6 Cluster clean Significant reduction with monomers As received 200eV clean 40 38 36 34 32 30 28 26 24 22
Benefits of cluster cleaning (Ta 2 O 5 ) In addition to reduced preferential sputtering: Increased photoelectron signal Removal of attenuating contamination Apparent improvement in energy resolution observed in Ta 4f region Removal of other Ta compounds (hydroxides, carbonates, etc) As received Cluster clean Improvement in resolution after cluster clean FWHM / ev As received 1.30 Cluster cleaned 1.09 40 38 36 34 32 30 28 26 24 22
Conclusions Preferential sputtering of TiO 2 (and Ta 2 O 5 ) are substantially reduced using an Ar + n cluster source. The effect of ion beam etching TiO 2 leads not only to preferential sputtering of O, but also a modification of the bonding or polarization of O atoms in the TiO 2 bulk structure. The normal quantification of TiO 2 using the Ti 2p envelope leads to an under estimation of the Ti atomic percentage, as loss peaks have to be taken into account. The use of Ar n + cluster sources has the potential to resolve (or substantially improve) XPS preferential sputtering problems for many important technological functional materials and thin films (oxides, chalcogenides etc,) hence, providing accurate composition information. Acknowledgements: Thanks to Prof. Ray Boxman, Tel Aviv University, Israel, for providing the TiO 2 thin films.